PyPI page
Home page
Author:
None
License:
MIT
Summary:
NVIDIA cuDNN Frontend — Python and C++ Graph API with SOTA attention (SDPA / Flash Attention), MoE grouped GEMM fusions, and FP8/MXFP8 kernels for Hopper and Blackwell GPUs.
Latest version:
1.23.0
Optional dependencies:
apache-tvm-ffi
|
cuda-python
|
nvidia-cutlass-dsl
|
torch
|
torch-c-dlpack-ext
Downloads last day:
118,273
Downloads last week:
762,083
Downloads last month:
3,255,546